Date: Sat, 10 Jan 1998 00:02:49 +1030 From: Mike Smith <mike@smith.net.au> To: daniel_sobral@voga.com.br Cc: mike@smith.net.au, hackers@FreeBSD.ORG Subject: Re: Device Driver Message-ID: <199801091332.AAA00656@word.smith.net.au> In-Reply-To: Your message of "Fri, 09 Jan 1998 11:35:33 -0300." <83256587.004F044C.00@papagaio.voga.com.br>
next in thread | previous in thread | raw e-mail | index | archive | help
> > > Ah! You are building an encrypting *router*. Everything becomes much > > much much more complicated. You have to maintain state for *all* of > > the connections whose datagrams you are routing. This is respectably > > nontrivial. > > [snip] > > > I have to be honest; it really sounds like you have embarked on a > > product without actually *designing* the damn thing first. > > Then it doesn't sound right... :-) The product exists, and it's not mine. > I'm just writing a driver for an encryption card that will be used by the > product. Anyway, thanks for all the help. It seems all my problems haven > been solved. Ok. I think I haven't understood what you're attempting to achieve. Anyway. > In the end, I decided using tsleep, and they'll be creating a kernel > process and accessing the device through it's normal interface. > > BTW, if I read the source code right, one "unit" of tsleep normally > corresponds to 1/128 seconds, and 1/1024 while profiling, is that right? Each tsleep() count corresponds to 1/hz seconds, where hz should be considered opaque. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199801091332.AAA00656>